| 06:27 | topi`> | jnettlet[m]: OK sounds good. Where do you keep your 4.9 branch? |
| 09:40 | jnettlet[m]> | topi`: https://github.com/SolidRun/linux-fslc/tree/solidrun-imx_4.9.x_1.0.0_ga |
| 09:43 | jnettlet[m]> | topi`: note that I have added the Kconfig option MXC_USE_VENDOR_DRIVERS If this is enabled then it builds NXP's ipu, fb, galcore, vpu, media drivers. If it is not set it includes all of NXPs patches for other drivers but only devices already in mainline, or not conflicting. |
| 10:00 | topi`> | I think we're running only mainlined stuff, and for h264 decoding I've grabbed some coda firmware blob that seems to work with gstreamer |
| 10:00 | topi`> | but definitely good to have the option to use NXP wares |
| 10:01 | Ke> | NNNNNOOOOOOOO |
| 14:12 | Ke> | https://github.com/MarvellEmbeddedProcessors/atf-marvell/commits/atf-v1.3-armada-17.10 btw there is new atf with fixes |
| 14:13 | Ke> | complete new release anyway |
| 14:16 | jnettlet[m]> | Ke: https://photos.app.goo.gl/DjvTZk7yk9ffaXtt2 |
| 14:16 | jnettlet[m]> | I still have the backplate on the 3d printer behind me. Will add that later. |
| 14:17 | jnettlet[m]> | power button is wired to the internal reset header |
| 14:17 | Ke> | =o) |
| 14:17 | Ke> | would love a proper backplate |
| 14:17 | jnettlet[m]> | 256GB samsung 960 nvme drive in the pcie 4x slot |
| 14:18 | jnettlet[m]> | I will post the stl file once I am done tweaking |
| 14:20 | Ke> | I guess I have to modify it a bit for my older borad |
| 14:20 | Ke> | board |
| 14:21 | Ke> | wonder what would happen, if I tried edk2-open-platform branch armada-upstream |
| 15:27 | jnettlet[m]> | Oh and Ke I swapped out the default CPU fan with one of these. http://www.frozencpu.com/products/21913/fan-259/Scythe_Mini_Kaze_40mm_x_10mm_Silent_Mini_Fan_-_3500RPM_SY124010L.html |
| 15:27 | jnettlet[m]> | huge difference in fan noise reduction |
| 15:28 | jnettlet[m]> | I can have the box sitting right next to me and can barely hear it running |
| 15:28 | jnettlet[m]> | that is with an open top. if you cover the top, it is basically dead silent |
| 15:30 | Ke> | jnettlet[m]: does the pwm thing actually work nowadays? |
| 15:30 | Ke> | and who controls |
| 15:30 | jnettlet[m]> | not in mainline |
| 15:31 | Ke> | mainline kernel? |
| 15:31 | jnettlet[m]> | but with that fan I don't ramp up and down. Just having it spinning constant |
| 15:31 | Ke> | ie. it is controlled by the kernel? |
| 15:31 | jnettlet[m]> | yes |
| 15:31 | jnettlet[m]> | I haven't delved into looking what needs to be down to have it controlled by the MC firmware |
| 18:47 | Ke> | jnettlet[m]: do I need some new gcc to compile the workarounds |
| 19:32 | jnettlet[m]> | Ke: workarounds for what? Sorry think I missed something |
| 19:41 | Ke> | spectre-meltdown |
| 19:41 | Ke> | then new atf etc. |
| 19:53 | jnettlet[m]> | I think the gcc patches are only for x86 at the moment |
| 19:53 | Ke> | yeah, thought so |
| 19:53 | Ke> | well, I'll try native build anyway now |
| 19:54 | Ke> | it's not like I would bother to compile gcc myself anyway |
| 19:54 | Ke> | the chances of self compiled gcc working are like 0 |
| 19:56 | Ke> | well apparently I can't as gcc-7 is not supported anyway |